3 papers
physics.ao-ph2026
Do AI weather models miss extremes?
Marvin Vincent Gabler, Roberto Molinaro, Niall Siegenheim +5
First-generation AI weather models are often reported to underperform at extremes, mostly in reanalysis-based evaluations of deterministic regression systems. We verify eleven phys…
cs.LG2026
Universal Diffusion-Based Probabilistic Downscaling
Roberto Molinaro, Niall Siegenheim, Henry Martin +4
We introduce a universal diffusion-based downscaling framework that lifts deterministic low-resolution weather forecasts into probabilistic high-resolution predictions without any…
cs.LG2025
EPT-2 Technical Report
Roberto Molinaro, Niall Siegenheim, Niels Poulsen +12
We present EPT-2, the latest iteration in our Earth Physics Transformer (EPT) family of foundation AI models for Earth system forecasting. EPT-2 delivers substantial improvements o…